Interest-aware content discovery in peer-to-peer social networks.

With the increasing popularity and rapid development of Online Social Networks (OSNs), OSNs not only bring fundamental changes to information and communication technologies, but also make extensive and profound impact on all aspects of our social life. Efficient content discovery is a fundamental challenge for large-scale distributed OSNs. However, the similarity between social networks and online social networks leads us to believe that the existing social theories are useful for improving the performance of social content discovery in online social networks. In this paper, we propose an interest-aware social-like peer-to-peer (IASLP) model for social content discovery in OSNs by mimicking ten different social theories and strategies. In the IASLP network, network nodes with similar interests can meet, help each other and co-operate autonomously to identify useful contents. The presented model has been evaluated and simulated in a dynamic environment with an evolving network. The experimental results show that the recall of IASLP is 20% higher than the existing method SESD while the overhead is 10% lower. The IASLP can generate higher flexibility and adaptability and achieve better performance than the existing methods.